CN110097966B - Information reminding method and device and terminal equipment - Google Patents
Information reminding method and device and terminal equipment Download PDFInfo
- Publication number
- CN110097966B CN110097966B CN201810089209.8A CN201810089209A CN110097966B CN 110097966 B CN110097966 B CN 110097966B CN 201810089209 A CN201810089209 A CN 201810089209A CN 110097966 B CN110097966 B CN 110097966B
- Authority
- CN
- China
- Prior art keywords
- information
- fundus image
- time
- monitoring data
- interval
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Eye Examination Apparatus (AREA)
Abstract
The invention provides an information reminding method, an information reminding device and terminal equipment, wherein the information reminding method comprises the following steps: acquiring health data of a target object, generating fundus image expression progress information of the target object by using a preset incidence relation model according to the health data, and feeding back reminding information to the target object according to the fundus image expression progress information, wherein the reminding information is used for indicating whether the target object needs to see a doctor in time. The scheme of the invention can realize the feedback of the development condition of the retinopathy to the diabetic retina patient based on the health data of the diabetic retina patient, remind the diabetic retina patient to see a doctor or continue to observe in time, and avoid the further deterioration of the state of an illness.
Description
Technical Field
The invention relates to the technical field of mobile health, in particular to an information reminding method, an information reminding device and terminal equipment.
Background
Diabetic retinopathy is a typical series of pathological changes caused by retinal microvascular damage caused by diabetes. In China, the prevalence of retinopathy in diabetic patients ranges from 24.7% to 37.5%. In recent years, with the prevalence of diabetes self-management and the maturity of diabetic retinopathy stage and lesion position detection algorithms based on fundus images, a means is provided for the accumulation of personalized physical sign data and diabetic retina data of patients. The quantitative association relationship between the physical sign parameters of the patient and the fundus image expression can be found by depending on a large amount of data, and the diabetic can be provided with timely early warning to remind the diabetic whether retinopathy occurs or not in the self-management of the diabetes according to the personalized data of the patient.
At present, the early warning of diabetic retinopathy is mostly the early warning aiming at the new diabetic retinopathy. For the patients with the confirmed diabetic retinopathy, a method for effectively reminding the condition development condition of the patients is lacked, so that adverse effects that the condition is further worsened but not found in time can occur.
Disclosure of Invention
The embodiment of the invention provides an information reminding method, an information reminding device and terminal equipment, which can realize the feedback of the development condition of retinopathy to a diabetic retina patient based on the health data of the diabetic retina patient, remind the diabetic retina patient to see a doctor or continue to observe in time and avoid the further deterioration of the state of an illness.
In a first aspect, an embodiment of the present invention provides an information reminding method, including:
acquiring health data of a target object;
generating fundus image expression progress information of the target object by using a preset incidence relation model according to the health data;
according to the fundus image performance progress information, reminding information is fed back to the target object;
the preset incidence relation model is used for representing incidence relation between health data and fundus image representation progress information, and the reminding information is used for indicating whether the target object needs to be in time for seeing a doctor or not.
Optionally, the health data comprises one or more of the following:
blood glucose monitoring data, blood pressure monitoring data, glycosylated hemoglobin HbAlc monitoring data, blood fat monitoring data, the interval time from the current time to the last eye fundus image examination time and diabetes course information.
Optionally, the generating, according to the health data, fundus image representation progress information of the target object by using a preset association relation model includes:
generating blood glucose and blood pressure characteristic information by using a preset deep learning model according to the blood glucose monitoring data and the blood pressure monitoring data;
preprocessing the HbAlc monitoring data, the blood fat monitoring data, the interval time between the current time and the last fundus image examination time and the diabetes course information respectively to obtain HbAlc characteristic information, blood fat characteristic information, interval time characteristic information and diabetes course characteristic information;
and generating the fundus image expression progress information by using a preset machine learning model according to the blood glucose and blood pressure characteristic information, the HbAlc characteristic information, the blood fat characteristic information, the interval time characteristic information and the course characteristic information.
Optionally, the preset deep learning model is obtained by training according to blood glucose monitoring data and blood pressure monitoring data of a predetermined number of sample patients.
Optionally, the preset machine learning model is obtained by training according to blood glucose and blood pressure feature information, HbAlc feature information, blood lipid feature information, interval time feature information, and course feature information of a predetermined number of sample patients.
In a second aspect, an embodiment of the present invention further provides an information reminding apparatus, including:
the acquisition module is used for acquiring the health data of the target object;
the generation module is used for generating fundus image expression progress information of the target object by utilizing a preset incidence relation model according to the health data;
the reminding module is used for feeding back reminding information to the target object according to the fundus image performance progress information;
the preset incidence relation model is used for representing incidence relation between health data and fundus image representation progress information, and the reminding information is used for indicating whether the target object needs to be in time for seeing a doctor or not.
Optionally, the health data comprises one or more of the following:
blood glucose monitoring data, blood pressure monitoring data, glycosylated hemoglobin HbAlc monitoring data, blood fat monitoring data, the interval time from the current time to the last eye fundus image examination time and diabetes course information.
Optionally, the generating module includes:
the first generation unit is used for generating blood glucose and blood pressure characteristic information by using a preset deep learning model according to the blood glucose monitoring data and the blood pressure monitoring data;
the preprocessing unit is used for respectively preprocessing the HbAlc monitoring data, the blood fat monitoring data, the interval time between the current moment and the last fundus image examination moment and the diabetes course information to obtain HbAlc characteristic information, blood fat characteristic information, interval time characteristic information and diabetes course characteristic information;
and a second generating unit, configured to generate the fundus image expression progress information by using a preset machine learning model according to the blood glucose and blood pressure feature information, the HbAlc feature information, the blood lipid feature information, the interval time feature information, and the disease course feature information.
Optionally, the preset deep learning model is obtained by training according to blood glucose monitoring data and blood pressure monitoring data of a predetermined number of sample patients.
Optionally, the preset machine learning model is obtained by training according to blood glucose and blood pressure feature information, HbAlc feature information, blood lipid feature information, interval time feature information, and course feature information of a predetermined number of sample patients.
In a third aspect, an embodiment of the present invention further provides a terminal device, including a memory, a processor, and a computer program stored on the memory and capable of running on the processor, where the computer program implements the steps of the information reminding method when executed by the processor.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the above-mentioned information reminding method.
According to the information reminding method provided by the embodiment of the invention, the health data of the target object is obtained, the fundus image expression progress information of the target object is generated by utilizing the preset incidence relation model according to the health data, and the reminding information is fed back to the target object according to the fundus image expression progress information, wherein the reminding information is used for indicating whether the target object needs to see a doctor in time or not, so that the development condition of retinopathy can be fed back to the diabetic retina patient based on the health data of the diabetic retina patient, the diabetic retina patient is reminded to see a doctor in time or continue to observe, and the condition of the disease is prevented from further deteriorating.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive labor.
FIG. 1 is a flow chart of an information reminding method according to an embodiment of the present invention;
fig. 2 is a flowchart of generating fundus image presentation progress information according to the embodiment of the present invention;
FIG. 3 is a flow chart of modeling according to an embodiment of the present invention;
FIG. 4 is a flow chart of the generation of a reminder for a target patient according to an embodiment of the present invention;
FIG. 5 is a schematic structural diagram of an information reminding apparatus according to an embodiment of the present invention;
FIG. 6 is a second schematic structural diagram of an information reminding apparatus according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of a terminal device according to an embodiment of the present invention.
Detailed Description
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive labor.
Referring to fig. 1, an embodiment of the present invention provides an information reminding method, including the following steps:
step 101: health data of a target subject is acquired.
Wherein the health data may include one or more of the following in combination: blood glucose monitoring data, blood pressure monitoring data, glycosylated hemoglobin HbAlc monitoring data, blood lipid monitoring data, the interval time from the current time to the last eye fundus image examination time, diabetes course information and the like. The diabetes course information is specifically the interval time between the current time and the time when the diabetes is diagnosed.
In an embodiment of the present invention, the target object may be a diabetic retina patient. The blood glucose monitoring data may include fasting blood glucose series information, postprandial blood glucose series information, intra-day blood glucose standard deviation series information, and average blood glucose fluctuation amplitude series information obtained by daily monitoring at an interval from the current time to the last fundus image examination time. The blood pressure monitoring data may include systolic pressure sequence information and diastolic pressure sequence information monitored daily during an interval from the current time to the last fundus image examination time. The HbAlc monitoring data may be HbAlc obtained by examining blood of the target subject in an interval time from the present time to the last fundus image examination time. The blood lipid monitoring data may be blood lipid obtained by examining blood of the target subject at an interval from the present time to the last fundus image examination time.
Step 102: and generating fundus image expression progress information of the target object by using a preset incidence relation model according to the health data.
The preset incidence relation model is used for expressing incidence relation between health data and fundus image expression progress information and can be established in advance by a method combining deep learning and machine learning.
Step 103: and feeding back reminding information to the target object according to the fundus image expression progress information.
The reminding information is used for indicating whether the target object needs to be treated in time. The fundus image expression progress information here may take values of-1, 0, and 1, where-1 indicates that the fundus image is better in expression, 0 indicates that the fundus image is stable in expression (for example, no new or decreased suspected regions such as neovasculature, microaneurysms, hemorrhage, and exudation are present), and 1 indicates that the fundus image is deteriorated in expression (for example, the location of the newly added suspected neovasculature region, or suspected regions such as microaneurysms, hemorrhage, and exudation is changed, or the number of the suspected regions is increased, or the area of the suspected regions is increased). When the fundus image expression progress information is-1 and 0, a prompt message for coming to a hospital for a doctor in time can be fed back to the target object, and when the fundus image expression progress information is 1, a prompt message for continuing observation can be fed back to the target object.
According to the information reminding method provided by the embodiment of the invention, the health data of the target object is obtained, the fundus image expression progress information of the target object is generated by utilizing the preset incidence relation model according to the health data, and the reminding information is fed back to the target object according to the fundus image expression progress information, wherein the reminding information is used for indicating whether the target object needs to see a doctor in time or not, so that the development condition of retinopathy can be fed back to the diabetic object based on the health data of the diabetic retina patient, the diabetic retina patient is reminded to see a doctor in time or continue to observe, and the condition of the disease is prevented from further deteriorating.
In the embodiment of the present invention, referring to fig. 2, a process of generating fundus image representation progress information of a target object by using a preset association relation model according to health data of the target object may be:
step 21: and generating blood glucose and blood pressure characteristic information by using a preset deep learning model according to the blood glucose monitoring data and the blood pressure monitoring data.
Wherein, the preset deep learning model is obtained by training according to the blood sugar monitoring data and the blood pressure monitoring data of a preset number of sample patients. The preset deep learning model can be, for example, a preset convolutional neural network LeNet model, and is trained by using a deep learning method. The predetermined data may be selected according to the actual situation.
It should be noted that, in order to ensure that the preset deep learning model can accurately process the blood glucose monitoring data and the blood pressure monitoring data of the target object, the expression form of the blood glucose monitoring data and the blood pressure monitoring data of the target object should be consistent with the expression form of the blood glucose monitoring data and the blood pressure monitoring data of the sample patient for training the preset deep learning model.
Step 22: and respectively preprocessing HbAlc monitoring data, blood fat monitoring data, the interval time between the current time and the last fundus image examination time and the diabetes course information to obtain HbAlc characteristic information, blood fat characteristic information, interval time characteristic information and course characteristic information.
Step 23: and generating fundus image expression progress information by using a preset machine learning model according to the blood glucose and blood pressure characteristic information, the HbAlc characteristic information, the blood lipid characteristic information, the interval time characteristic information and the course characteristic information.
The preset machine learning model is obtained by training according to blood glucose and blood pressure characteristic information, HbAlc characteristic information, blood fat characteristic information, interval time characteristic information and course characteristic information of a preset number of sample patients. The preset machine learning model can be, for example, a preset random forest model or a preset Support Vector Machine (SVM) model, and is trained by a machine learning method.
It should be noted that, in order to ensure that the preset machine learning model can accurately generate the fundus image expression progress information of the target object, the expression forms of the blood glucose and blood pressure characteristic information, HbAlc characteristic information, blood lipid characteristic information, interval time characteristic information and disease course characteristic information of the target object should be consistent with the expression forms of the blood glucose and blood pressure characteristic information, HbAlc characteristic information, blood lipid characteristic information, interval time characteristic information and disease course characteristic information of the sample patient used for training the preset machine learning model.
The following describes an implementation process of a specific embodiment of the present invention with reference to fig. 3 and 4.
First, a deep learning LeNet model and a machine learning SVM model are established based on historical data of diabetic retinal patients (i.e., sample patients) in a database.
Before establishing the model, historical data of sample patients in the database is traversed, and the results of two adjacent fundus image examinations of each sample patient and health data in the process are recorded as a data unit. The change between the latter fundus image representation and the former fundus image representation is recorded as fundus image representation progress information Y, wherein the values of Y are-1, 0 and 1, -1 represents that the fundus image representation is improved, 0 represents that the fundus image representation is stable, and 1 represents that the fundus image representation is deteriorated.
For each data unit, resampling all fasting blood glucose values monitored by a sample patient daily to generate fasting blood glucose sequence information with the length of 32 dimensions, and recording as X1;
resampling all postprandial blood glucose values monitored by a sample patient daily to generate 32-dimensional postprandial blood glucose sequence information, and recording as X2;
calculating the Standard Deviation of Blood Glucose (SDBG) in a day for the date when at least two blood glucose data records exist, resampling all SDBG values monitored by a sample patient in a daily way, and generating SDBG sequence information with the length of 32 dimensions, wherein the SDBG sequence information is recorded as X3;
calculating average blood glucose fluctuation amplitude (MAGE, namely the average amplitude of residual effective blood glucose fluctuation after removing all blood glucose fluctuations of which the amplitude does not exceed 1 SDBG) for the date with at least two blood glucose data records, and resampling all MAGE values monitored by a sample patient daily to generate MAGE sequence information with the length of 32 dimensions, wherein the MAGE sequence information is recorded as X4;
resampling all systolic pressures monitored by a sample patient daily to generate systolic pressure sequence information with the length of 32 dimensions, and recording as X5;
all the routinely monitored diastolic pressures of the sample patient were resampled to generate a 32-dimensional diastolic pressure sequence information, denoted X6.
After the processing of the blood sugar and blood pressure data, a LeNet model is trained by using a deep learning method according to X1-X6 (input parameters) and corresponding Y (output parameters) of each data unit, and 84-dimensional features (X1 '-X84') of the last layer, i.e., a full connection layer, of the LeNet model are recorded as results of feature extraction on X1-X6, so that blood sugar and blood pressure feature information (X1 '-X84') is generated, as shown in fig. 3.
Referring to fig. 3 again, for each data unit, all hbalcs obtained by blood examination of the sample patient may be preprocessed, and a median of the hbalcs is calculated and recorded as HbAlc characteristic information X85';
preprocessing all blood lipids obtained by blood examination of a sample patient, calculating to obtain a median value of the blood lipids, and recording as blood lipid characteristic information X86';
calculating the interval time between two fundus image examinations, and recording the interval time as interval time characteristic information X87';
the course of diabetes (the time interval from the time of diabetes confirmation) of the sample patient in the subsequent fundus image examination was calculated and recorded as course characteristic information X88'.
After the processing of HbAlc, blood fat and the like, an SVM model can be trained by using a machine learning method according to X1 '-X88' (88 input parameters, numerical types) and corresponding Y (output parameters) of each data unit, the size of model parameters is determined, and the association relationship between X and Y is established. Wherein the size of the model parameters represents the importance of different input parameters for influencing Y.
Then, based on the established LeNet model and SVM model, reminding information is generated for the target patient (namely the target object).
It is noted that all health data of the target patient from the previous fundus image examination to date may be recorded as a new data unit before generating the reminder information for the target patient. Referring to fig. 4, when generating the reminder information for the target patient, the data in the new data unit may be preprocessed to generate blood sugar related series information X1-X4, blood pressure related series information X5-X6, HbAlc characteristic information X85 ', blood lipid characteristic information X86', interval time characteristic information X87 'and course characteristic information X88' by referring to the above-mentioned processing procedure of the health data of the sample patient; secondly, inputting the blood sugar related sequence information X1-X4 and the blood pressure related sequence information X5-X6 into the established LeNet model to generate blood sugar and blood pressure characteristic information X1 '-X84'; then, blood sugar and blood pressure characteristic information X1 '-X84', HbAlc characteristic information X85 ', blood fat characteristic information X86', interval time characteristic information X87 'and course characteristic information X88' are input into the established SVM model to generate fundus image expression progress information Y; finally, generating reminding information for the target patient according to the value of Y; for example, if Y is equal to 1, a prompt message for coming to a hospital for a doctor is fed back to the target patient, and if Y is equal to-1 or 0, a prompt message for continuing observation is fed back to the target patient.
The above embodiments describe the message reminding method of the present invention, and the message reminding device of the present invention will be described with reference to the embodiments and the accompanying drawings.
Referring to fig. 5, an embodiment of the present invention further provides an information reminding apparatus, including:
an obtaining module 51, configured to obtain health data of a target object;
a generating module 52, configured to generate, according to the health data, fundus image representation progress information of the target object by using a preset association relation model;
a reminding module 53, configured to feed back reminding information to the target object according to the fundus image performance progress information;
the preset incidence relation model is used for representing incidence relation between health data and fundus image representation progress information, and the reminding information is used for indicating whether the target object needs to be in time for seeing a doctor or not.
According to the information reminding device provided by the embodiment of the invention, the health data of the target object is acquired, the fundus image expression progress information of the target object is generated by utilizing the preset incidence relation model according to the health data, and the reminding information is fed back to the target object according to the fundus image expression progress information, wherein the reminding information is used for indicating whether the target object needs to see a doctor in time or not, so that the development condition of retinopathy can be fed back to the diabetic retina patient based on the health data of the diabetic retina patient, the diabetic retina patient is reminded to see a doctor in time or continue to observe, and the condition of the disease is prevented from further deteriorating.
Optionally, the health data comprises one or more of the following:
blood glucose monitoring data, blood pressure monitoring data, glycosylated hemoglobin HbAlc monitoring data, blood fat monitoring data, the interval time from the current time to the last eye fundus image examination time and diabetes course information.
In the embodiment of the present invention, referring to fig. 6, the generating module 52 includes:
the first generation unit 521 is configured to generate blood glucose and blood pressure characteristic information according to the blood glucose monitoring data and the blood pressure monitoring data by using a preset deep learning model;
a preprocessing unit 522, configured to respectively preprocess the HbAlc monitoring data, the blood lipid monitoring data, the interval between the current time and the previous fundus image examination time, and the diabetes course information, so as to obtain HbAlc characteristic information, blood lipid characteristic information, interval characteristic information, and course characteristic information;
a second generating unit 523, configured to generate the fundus image expression progress information by using a preset machine learning model according to the blood glucose and blood pressure feature information, the HbAlc feature information, the blood lipid feature information, the interval time feature information, and the disease course feature information.
Optionally, the preset deep learning model is obtained by training according to blood glucose monitoring data and blood pressure monitoring data of a predetermined number of sample patients.
Optionally, the preset machine learning model is obtained by training according to blood glucose and blood pressure feature information, HbAlc feature information, blood lipid feature information, interval time feature information, and course feature information of a predetermined number of sample patients.
The embodiment of the present invention further provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements each process of the above-mentioned information reminding method embodiment, and can achieve the same technical effect, and is not described herein again to avoid repetition.
Specifically, referring to fig. 7, the embodiment of the present invention further provides a terminal device, which includes a bus 71, a transceiver 72, an antenna 73, a bus interface 74, a processor 75, and a memory 76.
In this embodiment of the present invention, the terminal device further includes: a computer program stored on the memory 76 and executable on the processor 75, in particular, the computer program when executed by the processor 75 may implement the steps of:
acquiring health data of a target object;
generating fundus image expression progress information of the target object by using a preset incidence relation model according to the health data;
according to the fundus image performance progress information, reminding information is fed back to the target object;
the preset incidence relation model is used for representing incidence relation between health data and fundus image representation progress information, and the reminding information is used for indicating whether the target object needs to be in time for seeing a doctor or not.
In fig. 7, a bus architecture (represented by bus 71), bus 71 may include any number of interconnected buses and bridges, bus 71 linking together various circuits including one or more processors, represented by processor 75, and memory, represented by memory 76. The bus 71 may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface 74 provides an interface between the bus 71 and the transceiver 72. The transceiver 72 may be one element or a plurality of elements, such as a plurality of receivers and transmitters, providing a means for communicating with various other apparatus over a transmission medium. The data processed by the processor 75 is transmitted over a wireless medium via the antenna 73, and further, the antenna 73 receives the data and transmits the data to the processor 75.
The processor 75 is responsible for managing the bus 71 and general processing and may also provide various functions including timing, peripheral interfaces, voltage regulation, power management, and other control functions. And the memory 76 may be used to store data used by the processor 75 in performing operations.
Alternatively, the processor 75 may be a CPU, ASIC, FPGA or CPLD.
The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the computer program implements each process of the above-mentioned information reminding method embodiment, and can achieve the same technical effect, and is not described herein again to avoid repetition.
Computer-readable media, which include both non-transitory and non-transitory, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (8)
1. An information reminding method is characterized by comprising the following steps:
acquiring health data of a target object; wherein the target subject is a diabetic retina patient; the health data comprises one or more of the following in combination: blood glucose monitoring data in the interval between the current time and the last eye fundus image examination time, blood pressure monitoring data in the interval between the current time and the last eye fundus image examination time, glycosylated hemoglobin HbAlc monitoring data in the interval between the current time and the last eye fundus image examination time, blood lipid monitoring data in the interval between the current time and the last eye fundus image examination time, the interval between the current time and the last eye fundus image examination time and the interval between the current time and the diabetes mellitus confirmation time;
generating fundus image expression progress information of the target object by using a preset incidence relation model according to the health data; wherein the fundus image representation progress information is used for representing whether the representation of the fundus image from the last fundus image examination time to the current time is improved, stabilized or deteriorated;
according to the fundus image performance progress information, reminding information is fed back to the target object;
the preset incidence relation model is used for representing incidence relation between health data and fundus image expression progress information; when the fundus image performance progress information indicates that the performance of the fundus image from the last fundus image examination time to the current time improves or stabilizes, the reminder information instructs the target object to continue observation; alternatively, when the fundus image performance progress information indicates that the performance of the fundus image from the last fundus image examination time to the current time is deteriorated, the reminder information indicates that the target subject is in time to see a doctor.
2. The method according to claim 1, wherein the generating fundus image representation progression information of the target subject using a preset correlation model from the health data comprises:
generating blood glucose and blood pressure characteristic information by using a preset deep learning model according to the blood glucose monitoring data and the blood pressure monitoring data;
preprocessing the HbAlc monitoring data, the blood fat monitoring data, the interval time between the current time and the last fundus image examination time and the diabetes course information respectively to obtain HbAlc characteristic information, blood fat characteristic information, interval time characteristic information and diabetes course characteristic information;
and generating the fundus image expression progress information by using a preset machine learning model according to the blood glucose and blood pressure characteristic information, the HbAlc characteristic information, the blood fat characteristic information, the interval time characteristic information and the course characteristic information.
3. The method of claim 2, wherein the preset deep learning model is trained from blood glucose monitoring data and blood pressure monitoring data of a predetermined number of sample patients.
4. The method according to claim 2, wherein the preset machine learning model is trained based on fundus image expression progress information of a predetermined number of sample patients, and blood glucose blood pressure feature information, HbAlc feature information, blood lipid feature information, interval time feature information, and course feature information of the sample patients, which are related to the fundus image expression progress information.
5. An information reminder device, comprising:
the acquisition module is used for acquiring the health data of the target object; wherein the target subject is a diabetic retina patient; the health data comprises one or more of the following in combination: blood glucose monitoring data in the interval between the current time and the last eye fundus image examination time, blood pressure monitoring data in the interval between the current time and the last eye fundus image examination time, HbAlc monitoring data in the interval between the current time and the last eye fundus image examination time, blood lipid monitoring data in the interval between the current time and the last eye fundus image examination time, the interval between the current time and the last eye fundus image examination time and the interval between the current time and the time for confirming diabetes diagnosis;
the generation module is used for generating fundus image expression progress information of the target object by utilizing a preset incidence relation model according to the health data; wherein the fundus image representation progress information is used for representing whether the representation of the fundus image from the last fundus image examination time to the current time is improved, stabilized or deteriorated;
the reminding module is used for feeding back reminding information to the target object according to the fundus image performance progress information;
the preset incidence relation model is used for representing incidence relation between health data and fundus image expression progress information; when the fundus image performance progress information indicates that the performance of the fundus image from the last fundus image examination time to the current time improves or stabilizes, the reminder information instructs the target object to continue observation; alternatively, when the fundus image performance progress information indicates that the performance of the fundus image from the last fundus image examination time to the current time is deteriorated, the reminder information indicates that the target subject is in time to see a doctor.
6. The apparatus of claim 5, wherein the generating module comprises:
the first generation unit is used for generating blood glucose and blood pressure characteristic information by using a preset deep learning model according to the blood glucose monitoring data and the blood pressure monitoring data;
the preprocessing unit is used for respectively preprocessing the HbAlc monitoring data, the blood fat monitoring data, the interval time between the current moment and the last fundus image examination moment and the diabetes course information to obtain HbAlc characteristic information, blood fat characteristic information, interval time characteristic information and diabetes course characteristic information;
and a second generating unit, configured to generate the fundus image expression progress information by using a preset machine learning model according to the blood glucose and blood pressure feature information, the HbAlc feature information, the blood lipid feature information, the interval time feature information, and the disease course feature information.
7. A terminal device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the information alert method as claimed in any one of claims 1 to 4.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the information reminding method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810089209.8A CN110097966B (en) | 2018-01-30 | 2018-01-30 | Information reminding method and device and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810089209.8A CN110097966B (en) | 2018-01-30 | 2018-01-30 | Information reminding method and device and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110097966A CN110097966A (en) | 2019-08-06 |
| CN110097966B true CN110097966B (en) | 2021-09-14 |
Family
ID=67441950
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810089209.8A Active CN110097966B (en) | 2018-01-30 | 2018-01-30 | Information reminding method and device and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110097966B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114864093B (en) * | 2022-07-04 | 2022-11-01 | 北京鹰瞳科技发展股份有限公司 | Apparatus, method and storage medium for disease prediction based on fundus image |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106530295A (en) * | 2016-11-07 | 2017-03-22 | 首都医科大学 | Fundus image classification method and device of retinopathy |
| CN106709225A (en) * | 2015-11-12 | 2017-05-24 | 中国移动通信集团公司 | Early-warning method, device and system for diabetic retinopathy |
| CN106934798A (en) * | 2017-02-20 | 2017-07-07 | 苏州体素信息科技有限公司 | Diabetic retinopathy classification stage division based on deep learning |
| CN107203778A (en) * | 2017-05-05 | 2017-09-26 | 平安科技（深圳）有限公司 | PVR intensity grade detecting system and method |
| CN107330449A (en) * | 2017-06-13 | 2017-11-07 | 瑞达昇科技(大连)有限公司 | A kind of BDR sign detection method and device |
| CN107423571A (en) * | 2017-05-04 | 2017-12-01 | 深圳硅基仿生科技有限公司 | Diabetic retinopathy identifying system based on eye fundus image |
-
2018
- 2018-01-30 CN CN201810089209.8A patent/CN110097966B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106709225A (en) * | 2015-11-12 | 2017-05-24 | 中国移动通信集团公司 | Early-warning method, device and system for diabetic retinopathy |
| CN106530295A (en) * | 2016-11-07 | 2017-03-22 | 首都医科大学 | Fundus image classification method and device of retinopathy |
| CN106934798A (en) * | 2017-02-20 | 2017-07-07 | 苏州体素信息科技有限公司 | Diabetic retinopathy classification stage division based on deep learning |
| CN107423571A (en) * | 2017-05-04 | 2017-12-01 | 深圳硅基仿生科技有限公司 | Diabetic retinopathy identifying system based on eye fundus image |
| CN107203778A (en) * | 2017-05-05 | 2017-09-26 | 平安科技（深圳）有限公司 | PVR intensity grade detecting system and method |
| CN107330449A (en) * | 2017-06-13 | 2017-11-07 | 瑞达昇科技(大连)有限公司 | A kind of BDR sign detection method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110097966A (en) | 2019-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20240221161A1 (en) | Method and system for computer-aided triage | |
| US11636601B2 (en) | Processing fundus images using machine learning models | |
| US11488299B2 (en) | Method and system for computer-aided triage | |
| US20210269046A1 (en) | Estimator generation apparatus, monitoring apparatus, estimator generation method, and computer-readable storage medium storing estimator generation program | |
| CN109741804A (en) | A kind of information extracting method, device, electronic equipment and storage medium | |
| EP3850638B1 (en) | Processing fundus camera images using machine learning models trained using other modalities | |
| JP2022180466A (en) | Diagnosis support device, learning device, diagnosis support method, learning method, and program | |
| JP2006149654A (en) | Support of diagnosis about lesion of eye fundus | |
| CN110097966B (en) | Information reminding method and device and terminal equipment | |
| US20200126649A1 (en) | Data processing apparatus, data processing method, and data processing program | |
| US10621414B2 (en) | Methods and systems for image collection and dimensional analysis for medical diagnoses | |
| US20180256029A1 (en) | A method and system for automated healthcare monitoring of a patient | |
| CN111462067B (en) | Image segmentation method and device | |
| CN112100629A (en) | Medical data processing method, device and system | |
| JP2021069699A (en) | Information processing device, information processing method, and program | |
| Das Adhikari et al. | Computer Vision-Aided Diabetic Retinopathy Detection Using Cloud-Deployed Deep Learning Framework | |
| JP2016047076A (en) | Image processor, fundus photographing system, image processing method, and program | |
| JP2022089250A (en) | Information processing device, information processing method, program, and information processing system | |
| CN114391819A (en) | Health data processing method and device | |
| CN114680938A (en) | Carotid plaque measuring method and ultrasonic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |